Return To Abstract Listing

Efficacy, Education, Administration, Informatics

E2669. Contrast-Induced Nephropathy Prediction by Age, Glomerular Filtration, and Ejection Fraction Score

Krepsky A1,  Araujo G1,  Pivatto F1,  Polanczyk C2,  Wainstein M.1 1. Clinicas Hospital of Porto Alegre, Porto Alegre, Brazil; 2. Médicos do Corpo Clínico, Porto Alegre, Brazil

Address correspondence to A. Krepsky (akrepsky@hcpa.edu.br)

Objective: Contrast-induced nephropathy is a common event after exposure to contrast media. It is associated with increased morbidity, mortality, hospital length-of-stay and healthcare costs. Yet, it is still unclear whether contrast-induced nephropathy is a direct cause of major events or is merely a marker of high-risk patients. The strength of association between contrast-induced nephropathy and mortality varies among different studies, and a recent meta-analysis suggests that the relationship between contrast-induced nephropathy and clinical outcomes are substantially confounded by baseline clinical characteristics that predispose to both kidney injury and mortality. The main strategy to avoid contrast-induced nephropathy lies in saline hydration, since to date no pharmacologic prophylaxis has proved beneficial. In order to identify high-risk patients, several prediction models have been created. Age, creatinine, and ejection fraction score is a simple risk model developed to predict mortality in patients undergoing elective coronary artery bypass graft, with a similar or better predictive accuracy compared to more complex scores. The purpose of this study is to determine if a low complexity mortality risk model is able to predict contrast-induced nephropathy in patients undergoing intraarterial contrast media exposure.

Materials and Methods: This retrospective study included patients with acute myocardial infarction submitted to primary percutaneous coronary intervention (PCI) in a tertiary university hospital. Definition of contrast-induced nephropathy was a raise of 0.3 mg/dL or 50% in postprocedure (24–72 h) serum creatinine compared to baseline. Age, creatinine, and ejection fraction score was calculated as follows: (age/left ventricle ejection fraction) + 1 point was added for every 10 mL/min/1.73 m2 reduction in CrCl < 60 mL/min/1.73 m2 (up to a maximum of 6 points).

Results: We have included 347 patients between April 2011 and December 2015. Mean age was 60 years, and 65% were men. Baseline creatinine was 1.06 mg/dL. Mean contrast volume was 199 (± 92) mL and 13.3% developed contrast-induced nephropathy. In-hospital death occurred in 8.2% of these cases. In univariate analysis, age, diabetes, previous acetylsalicylic acid use, age, creatinine, and ejection fraction scores and Mehran scores were predictors of contrast-induced nephropathy. After multivariate adjustment, only age, creatinine, and ejection fraction scores and diabetes remained contrast-induced nephropathy predictors. Areas under the ROC curve of age, creatinine, and ejection fraction and Mehran were 0.733 (0.68–0.78) and 0.649 (0.59–0.70), respectively. When we compared both scores with DeLong test age, creatinine, and ejection fraction scores AUC was greater than Mehran’s (p = 0.03). A score of 2.33 or lower has a negative predictive value of 92.6% for contrast-induced nephropathy development.

Conclusion: Age, creatinine, and ejection fraction score is a simple user-friendly tool that was independently predictive of contrast-induced nephropathy in patients undergoing primary PCI in our study. Moreover, a low score had an excellent negative predictive value for contrast-induced nephropathy, which makes this complication unlikely in patients with a score < 2.33. This was developed to predict major cardiovascular outcomes but predicts contrast-induced nephropathy better than validated and well-known scores developed for this matter, although not in patients with acute myocardial infarction. Because pharmacologic prophylaxis remains controversial, the main strategy to avoid contrast-induced nephropathy lies in its prevention; therefore, identification of high-risk patients is essential.